Agonal Sequences in 14 Filmed Hangings With Comments - PHI ...

ORIGINAL ARTICLE
Agonal Sequences in 14 Filmed Hangings With Comments on the
Role of the Type of Suspension, Ischemic Habituation, and
Ethanol Intoxication on the Timing of Agonal Responses
Anny Sauvageau, MD, MSc,* Romano LaHarpe, MD,† David King, MD,‡ Graeme Dowling, MD,*
Sam Andrews, MD,§ Sean Kelly, MD, Corinne Ambrosi, MD, Jean-Pierre Guay, PhD,
and Vernon J. Geberth, MS, MPS for the Working Group on Human Asphyxia
Abstract: The Working Group on Human Asphyxia has analyzed 14 filmed
hangings: 9 autoerotic accidents, 4 suicides, and 1 homicide. The following
sequence of agonal responses was observed: rapid loss of consciousness in
10 3 seconds, mild generalized convulsions in 14 3 seconds, decerebrate
rigidity in 19 5 seconds, beginning of deep rhythmic abdominal respiratory
movements in 19 5 seconds, decorticate rigidity in 38 15 seconds,
loss of muscle tone in 1 minute 17 seconds 25 seconds, end of deep
abdominal respiratory movements in 1 minute 51 seconds 30 seconds, and
last muscle movement in 4 minutes 12 seconds 2 minutes 29 seconds. The
type of suspension and ethanol intoxication does not seem to influence the
timing of the agonal responses, whereas ischemic habituation in autoerotic
practitioner might decelerate the late responses to hanging.
Key Words: hanging, asphyxia, video recording, pathophysiology, human,
ethanol
(Am J Forensic Med Pathol 2010;XX: 000–000)
Hanging is a form of asphyxia characterized by a constriction of the
neck by a ligature tightened by the gravitational weight of the body
or part of the body. 1 Death has traditionally been attributed to 3 possible
mechanisms: closure of the blood vessels of the neck, compression of
the air passages, and vagal stimulation by pressure on the baroreceptors
in the carotid sinuses and the carotid body. 2–4 This explanation of the
mechanism of death is largely based on old writings and experimentation
from the end of the 19th century and beginning of the 20th. 5 Apart
from a few animal studies that gave very limited information on the
pathophysiology of hanging in human, 6 there was slight new development
on this issue until the recent studies on filmed hangings by the
Working Group on Human Asphyxia. 7,8
In a series of 8 filmed hanging, it was demonstrated that all
cases demonstrated deep rhythmic abdominal respiratory movements.
8 These respiratory movements were not only visualized but
were also clearly audible. This fact strongly challenges the theory of
the obstruction of the air passage as a mechanism of death in
hanging. It could be argued, however, that hearing breath sounds
Manuscript received June 7, 2010; accepted June 22, 2010.
From the *Office of the Chief Medical Examiner, Edmonton, Alberta, Canada,
†Centre Universitaire Romand de Médecine Légale—Site Genève, Unité de
Médecine Forensique, Geneva, Switzerland; ‡Regional Forensic Pathology
Unit, Hamilton General Hospital, Hamilton, Ontario, Canada; §Office of the
Chief Medical Examiner, Calgary, Alberta, Canada; Office of the Chief
Medical Examiner, Jamaica, NY; School of Criminology, Faculty of Arts and
Sciences, University of Montréal and Montreal Philippe-Pinel Institute, Montreal,
Quebec, Canada; and **P.H.I. Investigative Consultant Inc, Garnerville,
NY.
Correspondence: Anny Sauvageau, MD, MSc, Office of the Chief Medical
Examiner, 7007, 116 St, Edmonton, Alberta T6H 5R8, Canada. E-mail:
anny.sauvageau@gmail.com.
Copyright © 2010 by Lippincott Williams & Wilkins
ISSN: 0195-7910/10/0000-0001
DOI: 10.1097/PAF.0b013e3181efba3a
does not eliminate the possibility of a partial airway obstruction.
Although it is now clear that tracheal occlusion is not complete in
some hangings, it would be premature to totally exclude some
implication of partial airways obstruction in the mechanism of death
in all hangings.
Besides the presence of clear audible respiration in some
cases, the study of filmed hangings has also documented the sequence
of agonal responses in these deaths 8 : a rapid loss of consciousness
after 8 to 18 seconds, closely followed by convulsions at
10 to 19 seconds, then a complex pattern of decerebrate rigidity and
decorticate rigidity, a loss of muscle tone after 1 minute 38 seconds
to 2 minute 15 seconds, and isolated body movements until the last
movements between 1 minute 2 seconds and 7 minute 31 seconds.
The description of this agonal sequence constitutes a major advance
in the understanding of the pathophysiology of human hanging.
Nevertheless, additional cases are needed before we can pretend to
fully grasp the agonic sequence in hanging deaths and appreciate its
variability. This fact was recently emphasized by the report of an
unusual case of hanging without decerebrate or decorticate rigidity. 9
In this film of a hanging with complete suspension of the body, a
52-year-old man stepped off a stool and the movement of the body
stepping off the stool created a rotary movement around the ceiling’s
ring. Apart from the rolling around the ceiling’s ring, the body
stayed motionless for the duration of the movie, without any evidence
of decerebrate or decorticate rigidity. A possible elucidation
of this singularity was proposed: the particular revolving movement
of the body around the ceiling’s ring created a vestibular stimulation
that interfered with the development of the decerebrate rigidity. 9
Here we present 5 additional filmed hangings, raising the
number of filmed hangings studied to 14.
MATERIALS AND METHODS
For each video, we evaluated the time frame of the following
body responses: loss of consciousness, convulsions, decorticate
rigidity, decerebrate rigidity, loss of muscle tone, last muscle movement,
and respiratory responses. All recording were evaluated
jointly by at least 2 judges, one of the judges having seen all the
videos.
RESULTS
The Working Group on Human Asphyxia has compiled and
analyzed 14 filmed hangings: 9 autoerotic accidents, 4 suicides, and
1 homicide. All victims were adults. Most were white males, with
the exception of an Asiatic male (case 9) and a white female (case
11). A brief description of each case and the observed agonal
sequence are presented in Table 1. The time to the various responses
is assessed considering time zero to represent the onset of the final
hanging. In case 14, however, the victim tied the ligature very tightly
2 seconds before the hanging and in this case, the time estimates
were done considering time 0 to be the tightening of the ligature.
Am J Forensic Med Pathol Volume XX, Number X, XXX 2010 www.amjforensicmedicine.com | 1

Sauvageau et al Am J Forensic Med Pathol Volume XX, Number X, XXX 2010
TABLE 1. Agonal Responses in 14 Filmed Hangings
Respiratory
Movement Responses
Responses (Very
Deep Respiratory
Decorticate Loss of Last Attempt)
Loss of
Decerebrate Decorticate Rigidity 2 Muscle Muscle
No Circumstances Type of Suspension Toxicology Consciousness Convulsions Rigidity Rigidity 1 (and 3) Tone Movement Start End
1 Suicide Incomplete; Feet on the No intoxication 13 s 15 s 19 s 21 s 1 min 11 s 1 min 38 s 4 min 10 s 20 s 2 min 00 s
ground
2 Autoerotic Incomplete; feet on the No intoxication Nd 14 s 19 s 1 min 08 s 1 min 32 s 2 min 15 s 2 min 47 s 21 s 2 min 03 s
ground
3 Autoerotic Complete No intoxication 18 s 19 s 21 s 1 min 00 s 1 min 04 s 2 min 04 s 3 min 01 s 22 s 2 min 04 s
4* Suicide Incomplete; kneeling on No intoxication Nd 18 s Nd Nd Nd Nd Nd 24 s Nd
the ground
5 Autoerotic Incomplete; almost
Nd 10 s 13 s 1 min 19 s 59 s — 1 min 52 s 7 min 31 s 13 s 2 min 05 s
completely lying down
in a prone position
6 Autoerotic Incomplete; feet on the
Nd 8 s 11 s 31 s 33 s — — 1 min 02 s 19 s 1 min 02 s
ground
7 Autoerotic Incomplete, feet on the
Nd 10 s 10 s 11 s 26 s 34 s Nd
ground
†
Nd †
13 s Nd †
8 Autoerotic Incomplete; feet on the No intoxication 12 s 14 s 20 s 31 s — Nd
ground
†
Nd †
16 s Nd †
9 ‡
Suicide Complete Ethanol 10 mg/100 mL Nd Nd — — — — — Nd Nd
in the femoral blood
10 Autoerotic Seems incomplete; feet No intoxication 8 s 12 s 20 s 24 s 39 s (and 1 2 min 45 s 7 min 35 s 16 s 2 min 37 s
probably on the
min 55 s)
ground
11 Suicide Incomplete; feet on the No intoxication 8 s 11 s Nd
ground
§
43 s 57 s 1 min 25 s 3 min 15 s 32 s 1 min 15 s
12
Homicide
Incomplete, feet on the Ethanol 156 mg/100 mL 9 s Nd 16 s Nd Nd Nd Nd Nd Nd
ground
in the femoral
blood**; THC 6.5 ng/
mL in the femoral
blood
13 Autoerotic Complete Nd 10 s 15 s 22 s 39 s 50 s (and 1 min 53 s Nd 17 s 1 min 43 s
57 s)
14 Autoerotic Incomplete, feet on the
Nd 8 s 11 s 14 s 35 s 45 s (and 1 min 42 s Nd
ground
53 s)
†
Nd ††
Nd ††
*The film obtained from a surveillance camera was not of optimal quality and several responses are difficult to assess.
† A few copies of films have been cut to early to allow evaluation of the late responses.
‡ 9 There was no decerebrate or decorticate rigidities in this case (the case was published as a case report, ).
§ At the beginning of the hanging, both arms are bent with the fingers slide between the ligature and the neck. The decerebrate rigidity is not seen probably because of this restriction limb movement. Later, the right fingers
slip from the ligature and decorticate rigidity is seen.
Several body responses are difficult to assess because the victim is in the shadow.
An erotic asphyxia practitioner asked for the help of his spouse in his practice. He was hanging himself and indicating her when to stop the process. On a filmed hanging, she was seen to ignore his sign to stop the hanging
2 | www.amjforensicmedicine.com © 2010 Lippincott Williams & Wilkins
and let him die.
**The victim is seen consuming several beers at the beginning of the film, before the hanging.
†† The respiratory attempt are seen but the timing was difficult to assess because of the angle of the camera.
Nd indicates no data; —, not observed.

Am J Forensic Med Pathol Volume XX, Number X, XXX 2010 Agonal Sequences in 14 Filmed Hangings
Loss of consciousness was largely assessed by a close examination
of the victim’s face, voluntary movements, and body tonus.
In 3 cases (cases 2, 4, and 9), loss of consciousness was not possible
to evaluate: in case 2, the victim’s face was masked with underwear;
in case 4, image quality from the surveillance camera was not
optimal enough to estimate this issue adequately; in case 9, the onset
of the loss of consciousness and of convulsions was not clearly
identifiable. In all the other cases, a rapid loss of consciousness was
observed in 8 to 18 seconds (average, 10 3 seconds). The loss of
consciousness was closely followed by mild generalized, tonicclonic
convulsions in 10 to 19 seconds (average, 14 3 seconds).
Decerebrate rigidity then followed. This postural attitude of
marked extensor rigidity is characterized by a full extension of the
upper and lower limbs, with extension of the hips and knees,
adduction of the legs, internal rotation of the shoulders, extension of
the elbows, hyperpronation of the distal parts of the upper limbs with
finger extension at the metacarpophalangeal joints, and flexion at the
interphalangeal joints. This extensor posturing was observed in 11
seconds to 1 minute 19 seconds (average, 25 19 seconds). If we
exclude the only case in which decorticate rigidity preceded
decerebrate rigidity (case 5), the posturing occurred in an average of
19 5 seconds.
After the decerebrate rigidity, a decorticate rigidity occurred:
this postural attitude is characterized by marked extensor rigidity of
the legs, identical to the one observed in decerebrate rigidity, but
combined with rigidity of the flexors of the arms: the arms are flexed
and bent on the chest, with the hands clenched into fists. A first
phase of decorticate rigidity was relatively sudden and quick. It was
generally followed by a second and sometimes a third phases of
decorticate rigidity, these subsequent ones developing more slowly
and being more sustained. The first decorticate rigidity appeared in
21 seconds to 1 minute 8 seconds (average, 40 16 seconds). If we
exclude case 5, the posturing first begins in an average of 38 15
seconds.
Between 1 minute 38 seconds and 2 minutes 45 seconds
(average, 1 minute 17 seconds 25 seconds), the body lost its
muscle tone and became progressively flaccid. Subsequently, isolated
body movements were observed from time to time, the last one
occurring between 1 minute 2 seconds and 7 minutes 35 seconds
(average, 4 minutes 12 seconds 2 minutes 29 seconds).
As for the respiratory responses, deep rhythmic abdominal
respiratory movements were seen These respiratory movements,
with rhythmic rocking of the body by the contraction of the diaphragm,
started between 13 and 32 seconds (average, 19 5
seconds) and stopped between 1 minute 2 seconds and 2 minutes 37
seconds (average, 1 minute 51 seconds 30 seconds). It is worth
emphasizing that these respiratory movements were not only seen
but also heard when sound was available, confirming the passage of
air in the airways despite the hanging process.
DISCUSSION
Despite the great diversity of hangings included in this study,
the similarity between most cases is striking. The agonic sequence
observed in hanging is summarized in Table 2. This study provides
a comprehensive portrait of the pathophysiology of human hanging.
The Role of the Type of Suspension on the Timing
of Agonal Responses
A comparison of time delay for agonal responses in complete
suspension (cases 3 and 13) and incomplete suspension (cases 1–2,
4–8, 10–12, 14) do not reveal impressive differences. However, the
small number of cases in the complete suspension group precludes
statistical analyses. These preliminary results suggest that the type of
suspension may not be an important factor in the timing of agonal
TABLE 2. Agonal Sequence in Hanging
Loss of consciousness
Average Time
10 3s
Convulsions 14 3s
Decerebrate rigidity 19 5s
Stat of deep rhythmic abdominal respiratory
movements
19 5s
Decorticate rigidity 38 15 s
Loss of muscle tone 1 min 17 s 25 s
End of deep rhythmic abdominal respiratory
movements
1 min 51 s 30 s
Last muscle movement 4 min 12 s 2 min 29 s
TABLE 3. Timing of the Agonal Responses: Comparison of
Autoerotic Hangings Versus Nonautoerotic Hangings
Autoerotic Hangings
Nonautoerotic
Hangings
Loss of consciousness 11 3s 103s Convulsions 13 3s 154s Decerebrate rigidity 20 6s 182s Stat of deep rhythmic
abdominal respiratory
movements
17 3s* 256s* Decorticate rigidity 40 16 s 32 16 s
Loss of muscle tone 2 min 5 s 23 s* 1 min 31 s 9s*
End of deep rhythmic
abdominal respiratory
movements
1 min 56 s 31 s 1 min 38 s 32 s
Last muscle movement 4 min 23 s 2 min 59 s 3 min 42 s 39 s
*Comparisons between both groups using Mann-Whitney U were significant at P
0.05.
responses and therefore in the time to irreversible damage and death.
Further cases are needed to assess the validity of this observation.
In the present study, most cases of incomplete hangings were in
an upright position with feet on the ground, with only 1 case of
incomplete hanging in a kneeling position (case 4) and one case of
incomplete hanging lying down (case 5). By comparing the timing
of responses in these cases, there is no evidence that the position of the
incomplete suspension interferes with the time delay. Further cases
are needed to assess the validity of this preliminary statement. It
should be pointed out, however that the only case with an inversion
of the sequence of decerebrate and decorticate rigidity is the case in
a lying down position (case 5). It is unknown at this time if this
position plays a role in this inversion of the sequence.
The Role of Ischemic Habituation on the Timing of
Agonal Responses
Considering that autoerotic practitioners might develop over
time a certain ischemic habituation, it is theoretically possible that these
cases present a deceleration of the sequence. In contrast, as they often
play for a longer period with the hanging process before the final
hanging, it could be argue that on the contrary, their hanging sequence
will be accelerated. A comparison of the timing of agonal responses in
autoerotic hangings versus nonautoerotic hangings is presented in Table
3. Overall, the time delays for the early responses to hanging seem to be
relatively similar between both groups, with the exception of an accelerated
start of deep abdominal respiratory movements in the autoerotic
practitioners. As for the late responses to hanging, they seem to be
© 2010 Lippincott Williams & Wilkins www.amjforensicmedicine.com | 3

Sauvageau et al Am J Forensic Med Pathol Volume XX, Number X, XXX 2010
decelerated in autoerotic practitioners. Statistical analysis confirmed a
significant difference between both groups for 2 variables: the start of
the deep abdominal respiratory abdominal movement occurred significantly
faster in autoerotic practitioner whereas the loss of muscle tone
was significantly delayed. Further research is needed to determine
whether and under which circumstances ischemic habituation plays a
role on the agonal sequences.
The Role of Ethanol Intoxication on the Timing of
Agonal Responses
An expert-witness recently stated that a victim intoxicated by
ethanol will die more quickly from strangulation than a sober victim,
alleging that ethanol is a respiratory depressant. We could not find
any animal or human studies to support or refute this theory. In the
present study, 1 victim was intoxicated with ethanol (case 12). This
intoxication state does not seem to have accelerated the timing of
agonal responses. It would be interesting to further evaluate the
effect of ethanol on the timing of agonal responses to hanging on the
animal model of hanging previously proposed by Boghossian et al. 6
REFERENCES
1. Sauvageau A, Boghossian E. Classification of asphyxia: the need for standardization.
J Forensic Sci. In press.
2. DiMaio VJ, DiMaio D. Asphyxia Forensic Pathology. 2nd ed. Boca Raton, FL:
CRC Press; 2001:229–277.
3. Spitz WU. Asphyxia. In: Spitz WU, Spitz DJ, eds. Spitz and Fisher’s Medicolegal
Investigation of Death: Guidelines for the Application of Pathology to
Crime Investigation. 4th ed. Springfield, IL: Charles C Thomas Publisher LTD;
2006:783–845.
4. Saukko P, Knight B. Suffocation and “Asphyxia.” In: Saukko P, Knight B, eds.
Knight’s Forensic Pathology. 3rd ed. London, England: Arnold Publishers;
2004:352–367.
5. Clement R, Redpath M, Sauvageau A. Cause of death in hanging: a historical
review of the evolution of pathophysiological hypotheses. J Forensic Sci. In
press.
6. Boghossian E, Clement R, Sauvageau A. Respiratory, circulatory and neurological
responses to asphyxia by hanging: a review of animal models.
J Forensic Sci. In press.
7. Sauvageau A. Agonal sequences in four filmed hangings: analysis of respiratory
and movement responses to asphyxia by hanging. J Forensic Sci. 2009;
54:192–194.
8. Sauvageau A, LaHarpe R, Geberth VJ; The Working Group on Human
Asphyxia. Agonal sequences in eight filmed hangings: analysis of respiratory
and movement responses to asphyxia by hanging. J Forensic Sci. In
press.
9. Sauvageau A, Kelly S, Ambrosi C. A filmed hanging without decerebrate and
decorticate rigidity: a case report and pathophysiological considerations. In
press.
4 | www.amjforensicmedicine.com © 2010 Lippincott Williams & Wilkins